Skip to content

Bose–Einstein Statistics

Bose–Einstein statistics is the quantum statistics of indistinguishable bosons. It allows any nonnegative number of particles to occupy the same one-particle mode and gives the thermal occupation factor

nˉi=1exp⁡[β(ϵi−μ)]−1.\bar n_i = \frac{1}{ \exp[\beta(\epsilon_i-\mu)]-1 }.

Historically, this statistics emerged in two steps. Bose used new counting for light quanta to derive Planck’s radiation law. Einstein then extended the same counting logic to an ideal gas of material particles and predicted the possibility of macroscopic ground-state occupation.

The central change from classical counting is that identical quantum particles do not carry observable individual labels. For bosons, exchanging two identical particles does not produce a new physical state. In modern notation, the exchange rule for a bosonic state is

P12∣Ψ⟩=∣Ψ⟩,P_{12}\lvert\Psi\rangle = \lvert\Psi\rangle,

where P12P_{12} swaps two particle slots. In a wavefunction representation with complete one-particle labels q1q_1 and q2q_2,

Ψ(q2,q1)=Ψ(q1,q2).\Psi(q_2,q_1) = \Psi(q_1,q_2).

This is the modern exchange-symmetry statement. Historically, the same idea first appeared less abstractly as a change in statistical counting: count occupations of modes, not assignments of labeled particles to boxes.

The formal nonrelativistic rule is developed in Symmetrization Postulate. This historical page explains why such a rule became necessary.

Bosonic mode occupations can be

ni=0,1,2,….n_i=0,1,2,\ldots .

There is no Pauli exclusion restriction for identical bosons. If a one-particle mode aa is available, states with

∣0a⟩,∣1a⟩,∣2a⟩,…\lvert 0_a\rangle, \quad \lvert 1_a\rangle, \quad \lvert 2_a\rangle, \quad \ldots

are all allowed in the bosonic Fock-space description.

For example, three identical bosons in two modes aa and bb have occupation patterns

(3,0),(2,1),(1,2),(0,3).(3,0), \quad (2,1), \quad (1,2), \quad (0,3).

The labels describe mode occupations, not names of three hidden particles. This is why occupation-number language is more natural than slot-label language for many-boson systems.

Multiple occupancy should not be misread as a force. Bosons are allowed to share a mode, but whether they actually do so depends on the Hamiltonian, temperature, density, interactions, and constraints.

For blackbody radiation, photon number is not fixed. Photons can be emitted and absorbed by matter in the cavity walls. The chemical potential is therefore zero in the ordinary equilibrium blackbody problem:

μ=0.\mu=0.

The Bose–Einstein occupation factor becomes

nˉ(ν)=1exp⁡(hν/kBT)−1.\bar n(\nu) = \frac{1}{\exp(h\nu/k_{\mathrm B}T)-1}.

Together with the electromagnetic mode density, this gives Planck’s Radiation Law.

For an ideal gas of material bosons, particle number is fixed. The chemical potential μ\mu is then adjusted so that

N=∑i1exp⁡[β(ϵi−μ)]−1.N = \sum_i \frac{1}{ \exp[\beta(\epsilon_i-\mu)]-1 }.

The chemical potential must remain below the lowest single-particle energy:

μ<ϵ0.\mu<\epsilon_0.

At high temperature or low density, the occupation of every mode is small. Then

exp⁡[β(ϵi−μ)]≫1,\exp[\beta(\epsilon_i-\mu)]\gg1,

and the Bose–Einstein factor approaches the Maxwell–Boltzmann form

nˉi≈exp⁡[−β(ϵi−μ)].\bar n_i \approx \exp[-\beta(\epsilon_i-\mu)].

Classical statistics is therefore recovered as a dilute, nondegenerate limit rather than as the fundamental counting rule.

Einstein’s extension to material particles revealed a new possibility. In a three-dimensional ideal Bose gas, the excited states have a finite capacity at fixed temperature and density. When the total number of particles exceeds what the excited modes can hold, the surplus occupies the lowest mode macroscopically.

For a uniform ideal gas, the condensation condition is commonly expressed using the thermal de Broglie wavelength λT\lambda_T:

nλT3≳ζ(3/2).n\lambda_T^3 \gtrsim \zeta(3/2).

Here nn is number density. This formula is an ideal-gas benchmark, not a full description of interacting trapped condensates.

The historical point is striking: a change in counting led to the prediction of a qualitatively new state of matter. The experimental observation of dilute-gas Bose–Einstein condensation came much later, in 1995, after advances in laser cooling, evaporative cooling, magnetic trapping, and atomic detection.

Modernly, Bose–Einstein statistics is not only a thermal distribution. It is part of the structure of the Hilbert space for identical bosons. Fixed-NN bosonic states live in the symmetric subspace

Sym⁡Nh,\operatorname{Sym}^N\mathcal h,

and Fock space collects all particle numbers into a mode-occupation framework.

Examples include:

  • photons as bosonic field quanta;
  • phonons as bosonic quasiparticles in harmonic lattice theory;
  • helium-4 atoms as composite bosons in low-energy atomic physics;
  • dilute alkali atoms in ultracold-gas condensates;
  • bosonic lattice models such as the Bose-Hubbard model.

The spin-statistics theorem of relativistic quantum field theory explains why ordinary integer-spin particles are bosons under standard assumptions. Nonrelativistic quantum mechanics uses that connection as input through the symmetrization postulate.

  • Bose–Einstein statistics does not mean bosons must all occupy the same state.
  • Multiple occupancy is not an attractive force.
  • A condensate is not simply a very cold classical cloud.
  • Photons have μ=0\mu=0 in ordinary blackbody equilibrium because photon number is not conserved; material bosons generally need a chemical potential.
  • The Maxwell–Boltzmann distribution is a limit of Bose–Einstein statistics for dilute nondegenerate gases, not the exact rule for identical bosons.
  • The historical counting argument does not prove the relativistic spin-statistics theorem.
  • Composite particles behave as bosons only within the regime where their internal state is fixed and they can be treated as identical single objects.
  • S. N. Bose, “Plancks Gesetz und Lichtquantenhypothese,” Zeitschrift für Physik 26, 178-181, 1924.
  • A. Einstein, “Quantentheorie des einatomigen idealen Gases,” Sitzungsberichte der Preussischen Akademie der Wissenschaften, 261-267, 1924.
  • A. Einstein, “Quantentheorie des einatomigen idealen Gases. Zweite Abhandlung,” Sitzungsberichte der Preussischen Akademie der Wissenschaften, 3-14, 1925.
  • P. A. M. Dirac, “On the Theory of Quantum Mechanics,” Proceedings of the Royal Society A 112, 661-677, 1926, DOI: 10.1098/rspa.1926.0133.
  • K. Huang, Statistical Mechanics, 2nd ed., Wiley, 1987.
  • R. K. Pathria and P. D. Beale, Statistical Mechanics, 3rd ed., Academic Press, 2011.
  • L. Pitaevskii and S. Stringari, Bose-Einstein Condensation, Oxford University Press, 2003.
  1. List the possible occupation patterns for four identical bosons in two modes.
Solution

The occupations (na,nb)(n_a,n_b) must satisfy na+nb=4n_a+n_b=4 with nonnegative integers. The possibilities are

(4,0),(3,1),(2,2),(1,3),(0,4).(4,0), \quad (3,1), \quad (2,2), \quad (1,3), \quad (0,4).
  1. Show that the Bose–Einstein distribution reduces to the Maxwell–Boltzmann form when exp⁡[β(ϵi−μ)]≫1\exp[\beta(\epsilon_i-\mu)]\gg1.
Solution

Start from

nˉi=1exp⁡[β(ϵi−μ)]−1.\bar n_i = \frac{1}{ \exp[\beta(\epsilon_i-\mu)]-1 }.

If exp⁡[β(ϵi−μ)]≫1\exp[\beta(\epsilon_i-\mu)]\gg1, the −1-1 is negligible, so

nˉi≈exp⁡[−β(ϵi−μ)].\bar n_i \approx \exp[-\beta(\epsilon_i-\mu)].

This is the Maxwell–Boltzmann occupation factor up to the normalization controlled by μ\mu.

  1. Why is the chemical potential zero for blackbody photons but not generally zero for material bosons?
Solution

In blackbody equilibrium, photons can be created and destroyed by matter in the cavity walls, so photon number is not a conserved thermodynamic constraint. The corresponding chemical potential is zero. For a gas of material atoms, the particle number is fixed in the ideal-gas model, so μ\mu is adjusted to enforce that number.

  1. What does Bose–Einstein condensation mean in occupation-number language?
Solution

It means that one single-particle mode, usually the lowest-energy mode in the ideal-gas discussion, has occupation of order the total particle number:

N0=O(N).N_0=O(N).

This is macroscopic occupation of a mode, not the statement that every atom is a classical particle at rest.